Correlating preoperative cytology and colposcopic biopsy with final histology after conization of the cervix

Objectives. We set out to evaluate the predictive value of preoperative cytology and colposcopic histology in determining final pathology after conization of the cervix in the era of the Bethesda system. Methods. A retrospective review of the charts of 280 patients who had cold-knife conization was undertaken. Preoperative cytology and colposcopic biopsy histology were compared to the cone biopsy histology. The Spearman's correlation coefficient was used to determine the correlation between both preoperative cytology and histology and histology of the cone biopsy. Results. Preoperative cytology matched cone histology in only 56.5% of cases and was within one grade in 92.3% of cases, with a correlation of 0.369. Most outliers were high-grade cytology not confirmed at cone biopsy. Preoperative biopsy correlated even less with cone histology: Exact correlation occurred in 40% of cases, correlation within one grade was seen in 69.4% of cases, and the Spearman correlation coefficient was 0.328. Most outliers were high-grade lesions on cone biopsy that were not detected with colposcopic biopsy. Conclusions. Preoperative cytology and histology do not predict final pathology with enough accuracy to allow management to be deter-mined by them. Conization of the cervix remains a useful procedure in the triage of the abnormal cytological smear.